Kimi K3.1 Pro is Moonshot AI's most capable 3T-class open-weight model and the first release in the K3.1 series. It is a post-trained version of Kimi K3, trained with substantially more post-training compute, and is built for full-project coding, long-horizon agent work, research and professional knowledge work. On AI/ML API it is available through the OpenAI-compatible chat completions endpoint.
Figures reported by Moonshot AI at launch:
Full-project coding
Multi-file features, refactors and long-running engineering tasks across large repositories.
3D and game development
Moonshot AI positions the model for work in Blender, Unreal Engine, Unity, Godot and CAD tools.
Financial analysis and research
Financial modeling, document cross-referencing and investment research with long context.
Agentic workflows
Multi-step tool use where the model plans, calls tools and keeps context over long sessions.
Kimi K3.1 Pro is built for full-project coding, including multi-file features, refactors and long-horizon engineering tasks across large repositories.
Moonshot AI positions it for 3D modeling and game development, with direct work in Blender, Unreal Engine, Unity, Godot and CAD software.
It is suited to financial modeling, investment research and other knowledge work that requires cross-referencing many documents in a single context.
With reasoning, parallel tool calling and a 1M-token context window, it fits agentic workflows that run many steps with minimal supervision.
import requests r = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={"Authorization": "Bearer " + AIMLAPI_KEY}, json={ "model": "moonshot/kimi-k3-1-pro", "messages": [ { "role": "user", "content": "Hello!" } ] }, ) print(r.json())
const r = await fetch("https://api.aimlapi.com/v1/chat/completions", { method: "POST", headers: { Authorization: `Bearer ${process.env.AIMLAPI_KEY}`, "Content-Type": "application/json", }, body: JSON.stringify({ "model": "moonshot/kimi-k3-1-pro", "messages": [ { "role": "user", "content": "Hello!" } ] }), }); console.log(await r.json());
curl -X POST https://api.aimlapi.com/v1/chat/completions \ -H "Authorization: Bearer $AIMLAPI_KEY" \ -H "Content-Type: application/json" \ -d '{"model":"moonshot/kimi-k3-1-pro","messages":[{"role":"user","content":"Hello!"}]}'
OpenAI-compatible — swap the base URL and it works with your existing SDK.
| Type | Price |
|---|---|
| Input | |
| Output | |
| Cached input |
Cached input is billed at $0.20 / 1M tokens. The full 1M context window is included at standard pricing.
| Model | Input | Output | Context | Best for |
|---|---|---|---|---|
Kimi K3.1 Pro This page | Agentic coding & long-horizon work | |||
| Low-latency coding & agents | ||||
| Long-context, multimodal & agentic workflows | ||||
| Coding + agents | ||||
| Reasoning + agents |
Kimi K3.1 Pro is Moonshot AI's most capable 3T-class open-weight model and the first model in the K3.1 series. It is a post-trained version of Kimi K3, built for coding, research and agentic tool use, with a 1,048,576-token context window.
Kimi K3.1 Pro takes text and image input and returns text. It supports reasoning, tool calling (including parallel tool calls), vision, structured output and streaming.
Kimi K3.1 Pro has a 1,048,576-token context window, enough to keep large codebases, long documents and multi-step agent sessions in a single request.
Kimi K3.1 Pro costs $2 per 1M input tokens, $0.20 per 1M cached input tokens and $10 per 1M output tokens. Reasoning tokens are billed at the output rate.
Kimi K3.1 Pro keeps the Kimi K3 architecture (2.8T total / 104B active parameters, 1M context) and adds substantially more post-training. Moonshot AI reports gains in long-horizon coding (61.2% vs 55.95% on Kimi Code Bench 3.0 Long-horizon), financial modeling and research, and a lower price than Kimi K3.
Send a request to the OpenAI-compatible chat completions endpoint https://api.aimlapi.com/v1/chat/completions with the model ID moonshot/kimi-k3-1-pro and your AI/ML API key.
